This invention relates to novel antibodies, in particular to antibodies directed against the CD3 antigen complex.
Antibodies, or immunoglobulins, comprise two heavy chains linked together by disulphide bonds and two light chains, each light chain being linked to a respective heavy chain by disulphide bonds in a xe2x80x9cYxe2x80x9d shaped configuration. The two xe2x80x9carmsxe2x80x9d of the antibody are responsible for antigen binding, and include regions where the polypeptide structure varies, these xe2x80x9carmsxe2x80x9d being termed Fabxe2x80x2 fragments (fragmentxe2x80x94antigenxe2x80x94binding) or F(abxe2x80x2)2 which represents two Fabxe2x80x2 arms linked together by disulphide bonds. The xe2x80x9ctailxe2x80x9d or central axis of the antibody contains a fixed or constant sequence of peptides and is termed the Fc fragment (fragmentxe2x80x94crystalline). The production of monoclonal antibodies was first disclosed by Kohler and Milstein (Kohler and Milstein, Nature, 256, 495-497 (1975)). Such monoclonal antibodies have found widespread use as diagnostic agents and also in therapy.
Each heavy chain has at one end a variable domain followed by a number of constant domains. Each light chain has a variable domain at one end and a constant domain at its other end, the light chain variable domain being aligned with the variable domain of the heavy chain and the light chain constant domain being aligned with the first constant domain of the heavy chain (CH1). The constant domains in the light and heavy chains are not involved directly in binding the antibody to antigen. The light chain constant domain and the CH1 domain of the heavy chain account for 50% of each Fabxe2x80x2 fragment.
The variable domains of each pair of light and heavy chains form the antigen binding site. The domains on the light and heavy chains have the same general structure and each domain comprises four framework regions, whose sequences are relatively conserved, connected by three complementarity determining regions (CDRs) (Kabat et al, Sequences of Proteins of Immunological Interest, U.S. Department of Health and Human Services (1987)). The four framework regions largely adopt a beta-sheet conformation and the CDRs form loops connecting, and in some cases forming part of, the beta-sheet structure. The CDRs are held in close proximity by the framework regions and, with the CORs from the other domain, contribute to the formation of the antigen binding site.
The human CD3 antigen consists of a minimum of four invariant polypeptide chains, which are non-covalently associated with the T-cell receptors on the surface of T-cells, and is generally now referred to as the CD3 antigen complex. It is intimately involved in the process of T-cell activation in response to antigen recognition by the T-cell receptors.
All CD3 monoclonal antibodies can be used to sensitise T-cells to secondary proliferative stimuli such as IL1 (interleukin 1) and IL2 (interleukin 2). In addition, certain CD3 monoclonal antibodies are themselves mitogenic for T-cells. This property is isotype dependent and results from the interaction of the CD3 antibody Fc domain with Fc receptors on the surface of accessory cells.
Rodent CD3 antibodies have been used to influence immunological status by suppressing, enhancing or re-directing T-cell responses to antigens. They therefore have considerable therapeutic potential in the human for use as an immunosuppressive agent, for example for the treatment of rejection episodes following the transplantation of renal, hepatic and cardiac allografts. However their value is compromised by two main factors. The first is the antiglobulin response evoked due to the xenogeneic nature of the antibody. The second is the xe2x80x9cfirst dosexe2x80x9d syndrome experienced by patients following the initial administration of the antibody. The symptoms, which range in severity from fever and chills to pulmonary edema, and which in rare cases can cause death, are caused by the elevated levels of circulating cytokines associated with CD3-antibody induced T-cell activation. This phenomenon requires the cross-linking of the CD3 antigen on the surface of T-cells to accessory cells through Fc receptors; such proliferation does not occur with F(abxe2x80x2)2 fragments of CD3 antibodies.
The first problem can be addressed by re-shaping or xe2x80x9chumanisingxe2x80x9d the variable region genes of antibodies and expressing them in association with relevant human constant domain genes. This reduces the non-human content of the monoclonal antibody to such a low level that an antiglobulin response is unlikely. Such a reshaped antibody with a binding affinity for the CD3 antigen complex is described in UK Patent Application No. 9121126.8 (published as GB 2249310A) and its equivalents (European Patent Application No. 91917169.4, Japanese Patent Application No. 516117/91 and U.S. patent application Ser. No. 07/862,543).
There remains however the problem of the first dose response when these antibodies are used in therapy. Aglycosylation of antibodies has been described to reduce their ability to bind to Fc receptors in vitro in some cases. However, it is not predictable that this will be true of all antibodies, particularly in vivo, and aglycosylation may result in the introduction into the antibody of novel and unpredictable properties including novel Fc binding characteristics causing other undesirable effects. It is also possible that other undesirable properties not associated with Fc binding may be introduced to the antibody.
Moreover, it is of course of vital importance that aglycosylation is not accompanied by the loss of certain desirable features of Fc binding in addition to the loss of the undesirable features such as those attributable to the first dose response.
It has now been found, however, that it is possible to produce aglycosylated CD3 antibodies of the IgG subclass which surprisingly retain their antigen binding specificity and immunosuppressive properties and yet do not induce T cell mitogenesis in vitro and induce a reduced level of cytokine release in vivo, whilst still maintaining some Fc binding ability.
Accordingly, the invention provides an aglycosylated IgG antibody having a binding affinity for the CD3 antigen complex.
The term aglycosylated is employed in its normal usage to indicate that the antibodies according to the invention are not glycosylated. Although the present invention can be applied to antibodies having a binding affinity for a non-human CD3 antigen complex, for example various other mammalian CD3 antigens for veterinary use, the primary value of the invention lies in aglycosylated antibodies having an affinity for the human CD3 antigen complex for use in the human and the following discussion is particularly directed to that context.
Further discussion of CD3 antigens is to be found in the report of the First International Workshop and Conference on Human Leukocyte Differentiation Antigens and description of various glycosylated antibodies directed against the CD3 antigen is also to be found in the reports of this series of Workshops and Conferences, particularly the Third and Fourth, published by Oxford University Press. Specific examples of such antibodies include those described by Van Ller et al., Euro. J. Immunol., 1987, 17, 1599-1604, Alegre et al., J. Immunol., 1991, 140, 1184, and by Smith et al., ibid, 1986, 16, 478, the last publication relating to the IgG1 antibody UCHT1 and variants thereof. However, of particular interest as the basis for aglycosylated antibodies according to the present invention are the CDRs contained in the antibodies OKT3 and YTH 12.5.14.2. The antibody OKT3 is discussed in publications such as Chatenaud et al., Transplantation, 1991, 51, 334 and the New England Journal of Medicine paper, 1985, 313, 339, and also in European Patent No. 0 018 795 and US Pat. No. 4,361,539. The antibody YTH 12.5.14.2 (hereinafter referred to as YTH 12.5) is discussed in publications such as Clark et al., European3J. Immunol., 1989, 19, 381-388 and reshaped YTH 12.5, antibodies are the subject of UK Patent Application No. 9121126.8 and its equivalents, this application describing in detail the CDRs present in this antibody.
Aglycosylated antibodies containing one or more of the CDRs described in the above application are of particular interest. Thus the antibodies of the invention preferably have at least one CDR selected from the amino acid sequences:
(a) Ser-Phe-Pro-Met-Ala (SEQUENCE ID NO. 1),
(b) Thr-Ile-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val-Lys-Gly (SEQUENCE ID NO. 2),
(c) Phe-Arg-Gln-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr (SEQUENCE ID NO. 3),
(d) Thr-Leu-Ser-Ser-Gly-Asn-Ile-Glu-Asn-Asn-Tyr-Val-His (SEQUENCE ID NO. 4),
(e) Asp-Asp-Asp-Lys-Arg-Pro-Asp (SEQUENCE ID NO. 5),
(f) His-Ser-Tyr-Val-Ser-Ser-Phe-Asn-Val (SEQUENCE ID NO. 6), and conservatively modified variants thereof.
The term xe2x80x9cconservatively modified variantsxe2x80x9d is one well known in the art and indicates variants containing changes which are substantially without effect on antibody-antigen affinity.
The CDRs are situated within framework regions of the heavy chain (for (a), (b) and (c)) and light chain (for (d), (e) and (f)) variable domains. The antibody also comprises a constant domain.
In a preferred embodiment the aglycosylated antibody has three CDRs corresponding to the amino acid sequences (a), (b) and (c) above or conservatively modified variants thereof and/or three CDRs corresponding to amino acid sequences (d), (e) and (f) or conservatively modified variants thereof, the heavy chain CDRs (a), (b) and (c) being of most importance.
A preferred aglycosylated antibody with a binding affinity for the CD3 antigen thus has a heavy chain with at least one CDR and particularly three CDRs selected from the amino acid sequences:
(a) Ser-Phe-Pro-Met-Ala (SEQUENCE ID NO. 1),
(b) Thr-Ile-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val-Lys-Gly (SEQUENCE ID NO. 2),
(c) Phe-Arg-Gln-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr (SEQUENCE ID NO. 3), and conservatively modified variants thereof, and/or a light chain with at least one CDR and particularly three CDRs selected from the amino acid sequences:
(d) Thr-Leu-Ser-Ser-Gly-Asn-Ile-Glu-Asn-Asn-Tyr-Val-His (SEQUENCE ID NO. 4),
(e) Asp-Asp-Asp-Lys-Arg-Pro-Asp (SEQUENCE ID NO. 5),
(f) His-Ser-Tyr-Val-Ser-Ser-Phe-Asn-Val (SEQUENCE ID NO. 6), and conservatively modified variants thereof.
Where an aglycosylated antibody according to the invention contains preferred CDRs as described hereinbefore it conveniently contains both one or more of the specified heavy chain CDRs and one or more of the specified light chain CDRs. The CDRs (a), (b) and (c) are arranged in the heavy chain in the sequence: framework region 1/(a)/framework region 2/(b)/framework region 3/(c)/framework region 4 in a leaderxe2x86x92constant domain (n-terminal to C-terminal) direction and the CDRs (d), (e) and (f) are arranged in the light chain in the sequence: framework region 1/(d)/framework region 2/(e)/framework region 3/(f)/framework region 4 in a leaderxe2x86x92constant domain direction. It is preferred, therefore, that where all three are present the heavy chain CDRs are arranged in the sequence (a), (b), (c) in a leaderxe2x86x92constant domain direction and the light chain CDRs are arranged in the sequence (d), (e), (f) in a leaderxe2x86x92constant domain direction.
It should be appreciated however, that aglycosylated antibodies according to the invention may contain quite different CDRs from those described hereinbefore and that, even when this is not the case, it may be possible to have heavy chains and particularly light chains containing only one or two of the CDRs (a), (b) and (c) and (d), (e) and (f), respectively. However, although the presence of all six CDRs defined above is therefore not necessarily required in an aglycosylated antibody according to the present invention, all six CDRs will most usually be present in the most preferred antibodies. A particularly preferred aglycosylated antibody therefore has a heavy chain with three CDRs comprising the amino acid sequences (a), (b) and (c) or conservatively modified variants thereof and a light chain with three CDRs comprising the amino acid sequences (d), (e) and (f) or conservatively modified variants thereof in which the heavy chain CDRs are arranged in the order (a), (b), (c) in the leader constant region direction and the light chain CDRs are arranged in the order (d), (e), (f) in the leader constant region direction.
The CDRs may be of different origin to the variable framework region and/or to the constant region and, since the CDRs will usually be of rat or mouse origin, this is advantageous to avoid an antiglobulin response in the human, although the invention does extend to antibodies with such regions of rat or mouse origin.
More usually the CDRs are either of the same origin as the variable framework region but of a different origin from the constant region, for example in a part human chimaeric antibody, or, more commonly, the CDRs are of different origin from the variable framework region.
The preferred CDRs discussed hereinbefore are obtained from a rat CD3 antibody. Accordingly, although the variable domain framework region can take various forms, it is conveniently of or derived from those of a rodent, for example a rat or mouse, and more preferably of or derived from those of human origin. One possibility is for the antibody to have a variable domain framework region corresponding to that in the YTH12.5 hybridoma although the constant region will still preferably be of or derived from those of human origin. However the antibody of the invention is preferably in the humanised form as regards both the variable domain framework region and as discussed further hereinafter, the constant region.
Accordingly, the invention further comprises an aglycosylated antibody which has a binding affinity for the human CD3 antigen and in which the variable domain framework regions and/or the constant region are of or are derived from those of human origin.
Certain human variable domain framework sequences will be preferable for the grafting of the preferred CDR sequences, since the 3-dimensional conformation of the CDRs will be better maintained in such sequences and the antibody will retain a high level of binding affinity for the antigen. Desirable characteristics in such variable domain frameworks are the presence of key amino acids which maintain the structure of the CDR loops in order to ensure the affinity and specificity of the antibody for the CD3 antigen, the lambda type being preferred for the light chain.
Human variable region frameworks which are particularly suitable for use in conjunction with the above CDRs have been previously identified in UK Patent Application No. 9121126.8. The heavy chain variable (V) region frameworks are those coded for by the human VH type III gene VH26.D.J. which is from the B cell hybridoma cell line 18/2 (Genbank Code: Huminghat, Dersimonian et al., Journal of Immunology, 139, 2496-2501). The light chain variable region frameworks are those of the human VLxcex type VI gene SUT (Swissprot code; LV6CSHum, Solomon et al. In Glenner et al (Eds), Amyloidosis, Plenum Press N.Y., 1986, p.449.
The one or more preferred CDRs of the heavy chain of the rat anti-CD3 antibody are therefore preferably present in a human variable domain framework which has the following amino acid sequence reading in the leaderxe2x86x92constant region direction, CDR indicating a CDR (a), (b) or (c) as defined hereinbefore, a conservatively modified variant thereof or an alternative CDR:-Glu-Val-Gln-Leu-Leu-Glu-Ser-Gly-Gly-Gly-leu-Val-Gln-Pro-Gly-Gly-Ser-Leu-Arg-Leu-Ser-Cys-Ala-Ala-Ser-Gly-Phe-Thr-Phe-Ser-/CDR/-Trp-Val-Arg-Gln-Ala-Pro-Gly-Lys-Gly-Leu-Glu-Trp-Val-Ser-/CDR/-Arg-Phe-Thr-Ile-Ser-Arg-Asp-Asn-Ser-Lys-Asn-Thr-Leu-Tyr-Leu-Gln-Met-Asn-Ser-Leu-Arg-Ala-Glu-Asp-Thr-Ala-Val-Tyr-Tyr-Cys-Ala-Lys-/CDR/-Trp-Gly-Gln-Gly-Thr-Leu-Val-Thr-Val-Ser-Ser (SEQUENCE ID NO. 7/CDR/SEQUENCE ID NO. 8/CDR/SEQUENCE ID NO. 9/CDR/SEQUENCE ID NO. 10).
In an aglycosylated antibody containing all three preferred CDRs, the heavy chain variable region comprises the following sequence:-Glu-Val-Gln-Leu-Leu-Glu-Ser-Gly-Gly-Gly-Leu-Val-Gln-Pro-Gly-Gly-Ser-Leu-Arg-Leu-Ser-Cys-Ala-Ala-Ser-Gly-Phe-Thr-Phe-Ser-Ser-Phe-Pro-Met-Ala-Trp-Val-Arg-Gln-Ala-Pro-Gly-Lys-Gly-Leu-Glu-Trp-Val-Ser-Thr-Ile-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val-Lys-Gly-Arg-Phe-Thr-Ile-Ser-Arg-Asp-Asn-Ser-Lys-Asn-Thr-Leu-Tyr-Leu-Gln-Met-Asn-Ser-Leu-Arg-Ala-Glu-Asp-Thr-Ala-Val-Tyr-Tyr-Cys-Ala-Lys-Phe-Arg-Gln-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr-Trp-Gly-Gln-Gly-Thr-Leu-Val-Thr-Val-Ser-Ser (SEQUENCE ID NO. 11).
Similarly, the one or more preferred CDRs of the light chain of the rat CD3 antibody are therefore preferably present in a human variable domain framework which has the following amino acid sequence reading in the leaderxe2x86x92constant region direction, CDR indicating a CDR (d), (e) and (f) as defined hereinbefore, a conservatively modified variant thereof or an alternative CDR:-Asp-Phe-Met-Leu-Thr-Gln-Pro-His-Ser-Val-Ser-Glu-Ser-Pro-Gly-Lys-Thr-Val-Ile-Ile-Ser-Cys-/CDR/-Trp-Tyr-Gln-Gln-Arg-Pro-Gly-Arg-Ala-Pro-Thr-Thr-Val-Ile-Phe-/CDR/-Gly-Val-Pro-Asp-Arg-Phe-Ser-Gly-Ser-Ile-Asp-Arg-Ser-Ser-Asn-Ser-Ala-Ser-Leu-Thr-Ile-Ser-Gly-Leu-Gln-Thr-Glu-Asp-Glu-Ala-Asp-Tyr-Tyr-Cys-/CDR/-Phe-Gly-Gly-Gly-Thr-Lys-Leu-Thr-Val-Leu (SEQUENCE ID NO. 12/CDR/SEQUENCE ID NO. 13/CDR/SEQUENCE ID NO. 14/CDR/SEQUENCE ID NO. 15).
In an aglycosylated antibody containing all three preferred CDRs the light chain variable region comprises the following sequence:-Asp-Phe-Met-Leu-Thr-Gln-Pro-His-Ser-Val-Ser-Glu-Ser-Pro-Gly-Lys-Thr-Val-Ile-Ile-Ser-Cys-Thr-Leu-Ser-Ser-Gly-Asn-Ile-Glu-Asn-Asn-Tyr-Val-His-Trp-Tyr-Gln-Gln-Arg-Pro-Gly-Arg-Ala-Pro-Thr-Thr-Val-Ile-Phe-Asp-Asp-Asp-Lys-Arg-Pro-Asp-Gly-Val-Pro-Asp-Arg-Phe-Ser-Gly-Ser-Ile-Asp-Arg-Ser-Ser-Asn-Ser-Ala-Ser-Leu-Thr-Ile-Ser-Gly-Leu-Gln-Thr-Glu-Asp-Glu-Ala-Asp-Tyr-Tyr-Cys-His-Ser-Tyr-Val-Ser-Ser-Phe-Asn-Val-Phe-Gly-Gly-Gly-Thr-Lys-Leu-Thr-Val-Leu (SEQUENCE ID NO. 16).
The variable domains, for example comprising one or more preferred CDRs as described above, preferably in the humanised form having human antibody-derived variable framework regions, are attached to appropriate constant domains.
The heavy and light chain constant regions can be based on antibodies of different types as desired subject to the antibody being an IgG antibody, but although they may be of or derived from those of rat or mouse origin they are preferably of or are derived from those of human origin. For the light chain the constant region is preferably of the lambda type and for the heavy chain it is preferably of an IgG isotype, especially IgG1, modified to effect aglycosylation as appropriate. All human constant regions of the IgG isotype are known to be glycosylated at the asparagine residue at position 297, which makes up part of the N-glycosylation motif Asparagine297-X298-Serine299 or Threonine299, where X is the residue of any amino acid except proline. The antibody of the invention may thus be aglycosylated by the replacement of Asparagine297 in such a constant region with another amino acid which cannot be glycosylated. Any other amino acid residue can potentially be used, but alanine is the most preferred. Alternatively, glycosylation at Asparagine297 can be prevented by altering one of the other residues of the motif, e.g. by replacing residue 298 by proline, or residue 299 by any amino acid other than serine or threonine. Techniques for performing this site directed mutagenesis are well known to those skiIled in the art and may for example be performed using a site directed mutagenesis kit such, for example, as that commercially available from Amersham. The procedure is further exemplified hereinafter.
It is well recognised in the art that the replacement of one amino acid in a CDR with another amino acid having similar properties, for example the replacement of a glutamic acid residue with an aspartic acid residue, may not substantially alter the properties or structure of the peptide or protein in which the substitution or substitutions were made. Thus, the aglycosylated antibodies of the present invention include those antibodies containing the preferred CDRs but with a specified amino acid sequence in which such a substitution or substitutions have occurred without substantially altering the binding affinity and specificity of the CDRs. Alternatively, deletions may be made in the amino acid residue sequence of the CDRs or the sequences may be extended at one or both of the N- and C-termini whilst still retaining activity.
Preferred aglycosylated antibodies according to the present invention are such that the affinity constant for the antigen is 105 molexe2x88x921 or more, for example up to 1012 molexe2x88x921. Ligands of different affinities may be suitable for different uses so that, for example, an affinity of 106, 107 or 108 molexe2x88x921 or more may be appropriate in some cases. However antibodies with an affinity in the range of 106 to 108 molexe2x88x921 will often be suitable. Conveniently the antibodies also do not exhibit any substantial binding affinity for other antigens. Binding affinities of the antibody and antibody specificity may be tested by assay procedures such as those described in the Examples section hereinafter, (Effector Cell Retargetting Assay), or by techniques such as ELLSA and other immunoassays.
Antibodies according to the invention are aglycosylated IgG CD3 antibodies having a xe2x80x9cYxe2x80x9d shaped configuration which may have two identical light and two identical heavy chains and are thus bivalent with each antigen binding site having an affinity for the CD3 antigen. Alternatively, the invention is also applicable to antibodies in which only one of the arms of the antibody has a binding affinity for the CD3 antigen. Such antibodies may take various forms. Thus the other arm of the antibody may have a binding affinity for an antigen other than CD3 so that the antibody is a bispecific antibody, for example as described in U.S. Pat. No. 4,474,893 and European Patent Applications Nos. 87907123.1 and 87907124.9. Alternatively, the antibody may have only one arm which exhibits a binding affinity, such an antibody being termed xe2x80x9cmonovalentxe2x80x9d.
Monovalent antibodies (or antibody fragments) may be prepared in a number of ways. Glennie and Stevenson (Nature, 295, 712-713, (1982)) describe a method of preparing monovalent antibodies by enzymic digestion. Stevenson et al. describe a second approach to monovalent antibody preparation in which enzymatically produced Fabxe2x80x2 and Fc fragments are chemically cross-linked (Anticancer Drug Design, 3, 219-230 (1989)). In these methods the resulting monovalent antibodies have lost one of their Fabxe2x80x2 arms. A third method of preparing monovalent antibodies is described in European Patent No. 131424. In this approach the xe2x80x9cYxe2x80x9d shape of the antibody is maintained, but only one of the two Fabxe2x80x2 domains will bind to the antigen. This is achieved by introducing into the hybridoma a gene coding for an irrelevant light chain which will combine with the heavy chain of the antibody to produce a mixture of products in which the monovalent antibody is the one of interest.
More preferably, however, the monovalent aglycosylated CD3 antibodies of the invention are prepared by the following method. This involves the introduction into a suitable expression system, for example a cell system as described hereinafter, together with genes coding for the heavy and light chains, of a gene coding for a truncated heavy chain in which the variable region domain and first constant region domain of the heavy chain are absent, the gene lacking the exon for each of these domains. This results in the production by the cell system of a mixture of (a) antibodies which are complete bivalent antibodies, (b) antibody fragments consisting only of two truncated heavy chains (i.e. an Fc fragment) and (c) fragments of antibody which are monovalent for the CD3 antigen, consisting of a truncated heavy chain and a light chain in association with the normal heavy chain. Such an antibody fragment (c) is monovalent since it has any only one Fabxe2x80x2 arm. Production of a monovalent antibody in the form of such a fragment by this method is preferred for a number of reasons. Thus, the resulting antibody fragment is easy to purify from a mixture of antibodies produced by the cell system since, for example, it may be separable simply on the basis of its molecular weight. This is not possible in the method of European Patent No. 131424 where the monovalent antibody produced has similar characteristics to a bivalent antibody in its size and outward appearance. Additionally, the production of a monovalent antibody fragment by the new method uses conditions which can more easily be controlled and is thus not as haphazard as an enzyme digestion/chemical coupling procedure which requires the separation of a complex reaction product, with the additional advantage that the cell line used will continue to produce monovalent antibody fragments, without the need for continuous synthesis procedures as required in the enzyme digestion/chemical coupling procedure.
It is believed that aglycosylated antibodies according to the invention do not occur in nature and these aglycosylated antibodies may in general be produced synthetically in a number of ways. Most conveniently, however, appropriate gene constructs for the constant and variable regions of the heavy and light chains which are present in the antibody are separately obtained and then inserted in a suitable expression system.
Genes encoding the variable domains of a ligand of the desired structure may be produced and conveniently attached to genes encoding the constant domains of an antibody which have undergone site directed mutagenesis. These constant genes may be obtained from hybridoma cDNA or from the chromosomal DNA and have undergone mutagenesis (site directed) to produce the aglycosylated constant regions. Genes encoding the variable regions may also be derived by gene synthesis techniques used in the identification of the CDRs contained herein. Suitable cloning vehicles for the DNA may be of various types.
Expression of these genes through culture of a cell system to produce a functional CD3 ligand is most conveniently effected by transforming a suitable prokaryotic or particularly eukaryotic cell system, particularly an immortalised mammalian cell line such as a myeloma cell line, for example the YB2/3.01/Ag20 (hereinafter referred to as Y0) rat myeloma cell, or Chinese hamster ovary cells (although the use of plant cells is also of interest), with expression vectors which include DNA coding for the various antibody regions, and then culturing the transformed cell system to produce the desired antibody. Such general techniques of use for the manufacture of ligands according to the present invention are well known in the very considerable art of genetic engineering and are described in publications such as xe2x80x9cMolecular Cloningxe2x80x9d by Sambrook, Fritsch and Maniatis, Cold Spring Harbour Laboratory Press, 1989 (2nd edition). The techniques are further illustrated by the Examples contained herein.
The present invention thus includes a process for the preparation of an aglycosylated IgG antibody having a binding affinity for the CD3 antigen which comprises culturing cells capable of expressing the antibody in order to effect expression thereof. The invention also includes a cell line which expresses an aglycosylated antibody according to the invention.
Preferred among such cell lines are those which comprise DNA sequences encoding the preferred CDRs described hereinbefore. A group of nucleotide sequences coding for the CDRs (a) to (f) described hereinbefore is as indicated under (a) to (f) below, respectively, but it will be appreciated that the degeneracy of the genetic code permits variations to be made in these sequences whilst still encoding for the CDRsxe2x80x2 amino acid sequences.
(a) AGCTTTCCAA TGGCC (SEQUENCE ID NO. 17)
(b) ACCATTAGTA CTAGTGGTGG TAGAACTTAC TATCGAGACT CCGTGAAGGG C (SEQUENCE ID NO. 18)
(c) TTTTCGGCAGT ACAGTGGTGG CTTTGATTAC (SEQUENCE ID NO. 19)
(d) ACACTCAGCT CTGGTAACAT AGAAAACAAC TATGTGCAC (SEQUENCE ID NO. 20)
(e) GATGATGATA AGAGACCGGA T (SEQUENCE ID NO. 21)
(f) CATTCTTATG TTAGTAGTTT TAATGTT (SEQUENCE ID NO. 22)
Such cell lines will particularly contain larger DNA sequences which comprise (1) DNA expressing human heavy chain variable framework regions and one or more of (a), (b) and (c), and (2) DNA expressing human light chain variable framework regions and one or more of (d), (e) and (f). A specific example of such DNA is that sequenced(1) indicated below which codes for the CDRs (a), (b) and (c) arranged in the heavy chain framework coded for by the human VH type III gene VH26.D.J. as discussed hereinbefore and that sequence (2) indicated below which codes for the CDRs (d), (e) and (f) arranged in the light chain framework coded for by the human VLxcex type VI gene SUT. The CDR sequences (a), (b), (c), (d), (e) and (f) have been underlined.
(1) GAGGTCCAAC TGCTGGAGTC TGGGGGCGGT TTAGTGCAGC CTGGAGGGTC CCTGAGACTC TCCTGTGCAG CCTCAGGATT CACTTTCAGT AGCTTTCCAA TGGCCTGGGT CCGCCAGGCT CCAGGGAAGG GTCTGGAGTG GGTCTCAACC ATTAGTACTA GTGGTGGTAGAACTTACTAT CGAGACTCCG TGAAGGGCCG ATTCACTATC TCCAGAGATA ATAGCAAAAA TACCCTATAC CTGCAAATGA ATAGTCTGAG GGCTGAGGAC ACGGCCGTCT AlTACTGTGC AAAATTTCGG CAGTACAGTG GTGGCTTTGA TTACTGGGGC CAAGGGACCC TGGTCACCG CTCCTCA (SEQUENCE ID NO. 23)
(2) GACTTCATGC TGACTCAGCC CCACTCTGTG TCTGAGTCTC CCGGAAAGAC AGTCATTATT TCTTGCACAC TCAGCTCTGG TAACATAGAA AACAACTATG TGCACTGGTA CCAGCAAAGG CCGGGAAGAG CTCCCACCAC TGTGATTTTC GATGATGATA AGAGACCGGA TGGTGTCCCT GACAGGTTCT CTGGCTCCAT TGACAGGTCT TCCAACTCAG CCTCCCTGAC AATCAGTGGT CTGCAAACTG AAGATGAAGC TGACTACTAC TGTCATTCTT ATGTTAGTAG TTTTAATGTT TTCGGCGGTG GAACAAAGCT CACTGTCCTT (SEQUENCE ID NO. 24)
The cell lines will of course also particularly contain DNA sequences expressing the heavy and light chain constant regions.
The humanised aglycosylated antibodies in accordance with the invention have therapeutic value. In particular, such aglycosylated antibodies, especially a humanised aglycosylated antibody with a specificity for the human CD3 antigen, has valuable applications in immunosuppression, particularly in the control of graft rejection, where it is especially desirable that immunosuppression is temporary rather than total, and thus that T-cells are not completely destroyed, but instead rendered non-functional by antibody blockade of the CD3 antigenxe2x80x94TCR complex. In addition, the aglycosylated CD3 antibodies may have potential in other areas such as in the treatment of cancer, specifically in the construction of bispecific antibodies (for effector cell retargetting) or antibody-toxin conjugates, where the efficacy of the therapeutic agent would be compromised by Fc-mediated killing of the effector cells or non-specific killing of Fc receptor bearing cells respectively.
In a further aspect, the invention thus includes a method of treating patients with cancer, particularly a lymphoma, or for immunosuppression purposes, for instance in a case where graft rejection may occur, comprising administering a therapeutically effective amount of an aglycosylated antibody in accordance with the invention.
Aglycosylated antibodies in accordance with the invention may be formulated for administration to patients by administering the said antibody together with a physiologically acceptable diluent or carrier. The antibodies are preferably administered in an injectable form together with such a diluent or carrier which is sterile and pyrogen free. By way of guidance it may be stated that a suitable dose of antibody is about 1-10 mg injected daily over a time period of, for example 10 days, although due to the elimination of the first dose response it will be possible if desired to adminster higher amounts of the antibody, for example even up to 100 mg daily, depending on the individual patient""s needs. Veterinary use is on a similar g/kg dosage basis.